基于极值理论的EXPAR时序模型异常点诊断

本文基于极值理论给出诊断EXPAR模型异常点检验统计量的渐近分布,并依此渐近分布来选取检验的临界值。这种方法选取的临界值可保证控制在一定显著性水平下,而且可以计算渐近p值,比仿真选取的临界值更科学合理。     

基于W,LR和LM检验的误差方差预检验估计的比较（英文）

本文研究多元t分布模型中误差方差分别基于W, LR和LM检验的预检验估计问题.首先,分别获得了基于三个大样本检验的预检验估计和对应的风险;其次,基于理论分析和数值分析的方法得到了各个估计的优良表现,结果表明预检验估计的风险在检验临界值为1时达到最小,并得到了各个估计保持最优的条件;最后,通过自助模拟的方法进一步分析了各个估计的优良性,所得结果和理论分析是一致的.     

基于Beta分布形状的拟合优度检验

本文指出了拟合优度检验中选用经验频率公式时存在的误区,对顺序统计量失效概率分布进行了偏态和峰态分析,提出了一个基于Beta分布形状特征的经验分布函数,给出了精确值和近似值两种计算方法,在此基础上建立新的极值型检验统计量．利用Monte Carlo方法进行数值模拟,得到0．01、0．05、0．1显著度水平下检验统计量的临界值,并利用常用的分布模型进行检验功效比较,数值模拟结果表明基于分布形状的经验频率公式能更好地反映顺序统计量失效概率分布的集中趋势,证明了本文提出的两种检验统计量在中小样本条件下具有更优的检验功效．     

测量误差模型参数变点的非参数检验

本讨论测量误差参数变点的检测问题,利用秩统计量,给出了模型只有一个变点的检验统计量,运用检验统计量渐近分布的性质,给出了一个计算检验淅近临界值的公式,由此我们可以较为客易计算检验的临界值。     

线性模型中M检验原假设分布的随机加权逼近

在线性模型中M-方法可以用于线性假设检验, 其中M检验、Wald检验和Rao的计分型检验是最常用的检验准则. 但是在计算这些检验的临界值时都涉及到未知参数的估计. 在本文中我们利用随机加权的方法来逼近这些检验的原假设分布. 结果表明在原假设和局部对立假设之下随机加权统计量的渐近分布与原检验统计量在原假设之下的渐近分布相同. 因此我们不需要对冗余参数进行估计,利用随机加权的方法就可以得到这些检验的临界值. 而且在局部对立假设之下可以实现对功效的计算. 当取不同的误差分布和不同的随机权时, 我们对本文的方法进行了蒙特卡洛模拟. 结果表明用随机加权方法来逼近原假设分布是非常精确的.     

变结构面板协整检验的响应面函数研究

Panel-ρ检验是常用的异质面板数据协整检验统方法,本文通过Monte Carlo模拟分析Panel-ρ协整检验方法的小样本性质,以及该方法对于变结构面板协整检验的检验功效.结果表明,在小样本情况下Panel-ρ统计量在原假设下的渐进分布会不同于该统计量的理论渐进分布,而且面板数据中存在的结构变化也会对渐进分布产生影响,从而降低Panel-ρ检验的检验功效.为了得到更加符合实际样本情况的统计量临界值,通过Monte Carlo模拟方法得到Panel-ρ协整检验方法的响应面函数,建立了统计量的临界值与面板数据的样本容量、结构变化类型的直接函数关系.Monte Carlo模拟检验表明,响应面函数法确实能够改善Panel-ρ协整检验的检验功效,在小样本容量和具有结构变化的情况下保证了面板协整检验的有效性.     

应用积分经验过程检验多元分布函数的相等性

本文引进投影积分经验过程用于检验两个或K个多元分布函数的相等性，自助法用于确定临界值的逼近，数论方法有效地计算自动法确定的临界值，且进行了一些模拟试验．     

方向数据的统计分析(Ⅴ)

4分布参数和模型参数的估计 本节我们对服从VonMises分布的方向数据,简单介绍有关的参数估计方法和思想, 从　　2开始,我们逐渐涉及到有关方向数据统计分析的各种图表.顺序排列于下。 附表1Von Mises分布表 附表 2 Von Mises分布合向量长度ρ= A(6) 附表3 Von Mise。分布K的LS估计R=A-1(R) 附表4检验流计量C的临界值 附表5检验流计量R的临界值 附表6 Von Mises分布的分位点 附表7 u0的置信区间表 附表8K的置信区间表 附表9 R=(R1; R1)/N的 5%临界值 本刊因篇幅所限,不刊登以上图表.需要者请与作者项静恬联系(邮政编码100080) 　4…     

测量误差模型只有一个变点的检验和估计

本文讨论了测量误差模型中参数只有一个变点的检验和估计问题，首先，给出其似然比检验统计量，然后，基于最小信息准则的原理，利用Schwarz信息准则（SIC），在多余参数已知和未知的情况下，分别给出了检验统计量，讨论了利用SIC方法给出的检验统计量的渐近分布，证明了基于似然比方法和SIC方法给出的变点估计是相同的，并且在一定条件下，给出了变点估计的极限分布，运用Monte-Carlo随机模拟的方法，分别给出了以上检验的临界值。     

长相依序列均值变点检测

研究了长相依序列均值变点检测问题.首先给出变点检验的Ratio统计量其次分别推导了原假设和备择假设下统计量的极限分布,最后用蒙特卡洛方法模拟出检验的临界值,并通过数值模拟和实例分析说明方法的有效性.     

Likelihood stabilization for ill-conditioned vector GARCH models

The likelihood of vector GARCH models is ill-conditioned because of two facts. First, when the series display high correlations, as often happens with financial data, some eigenvalues of the conditional covariance matrix are close to zero. Second, the likelihood function is very flat in the neighborhood of the optimum due to the functional form of the GARCH process. These facts explain the instability of multivariate GARCH estimation procedures. Building on this analysis, we suggest a data transformation which moves the critical eigenvalues far from zero and, therefore, improves the stability of iterative optimization methods. The transformed values are re-scaled principal components, so their interpretation is straightforward. The application of this technique is illustrated by modeling the short-run conditional correlations of four nominal exchange rates.      

Simultaneous prediction regions

Summary Two new methods for constructing simultaneous prediction regions are the subject of this article. Both methods simultaneously assert a collection of prediction regions, one prediction region for each future observable of interest. Both methods have the same aims: to control the overall coverage probability of the simultaneous prediction region and to keep equal the coverage probabilities of the individual prediction statements that make up the simultaneous region. The latter property is called balance.The two approaches differ in their choice of critical values. For leading cases, the first method achieves the desired overall coverage probability and the desired balance up to errors of ordern ^–1, wheren is the size of the learning sample. The second method reduces both errors to ordern ^–2. Calculating critical values in the second approach usually relies on a bootstrap algorithm.If overall coverage probability and degree of balance are instead calculatedconditionally given the learning sample, the two methods show the same asymptotic performance. This result reflects intrinsic limits on the extent to which conditional coverage probabilities can be controlled in prediction.This research was supported in part by NSF Grant DMS-87-01426. Part of the work was done while the author was a guest of Sonderforschungsbereich 123 at Universität Heidelberg     

基于二维t分布的条件VaR研究

根据多元t分布的定义及性质,推导出二维t分布随机变量差的条件分布仍服从t分布.在假定股票价格对数与收益率服从二维t分布的基础上,利用该性质,可以得到不同股票价格水平条件下,收益率的一维条件t分布,进而计算出价格条件VaR.利用多元t分布研究价格条件的收益率分布问题,与正态分布相比,较好地刻画了证券收益率分布的尖峰厚尾现象.     

光伏发电投资决策的实证分析——基于ARMA-GARCH模型修正后的动态规划法

运用可靠方法评估项目的最优临界值和最大机会价值是光伏发电投资决策面临的关键问题。本研究选取了与某光伏企业发电投资项目价值“孪生”的一只股票的836个日收盘价格(从2012年1月4日至2015年6月24日)建立波动率预测模型,并在此基础上修正了该项目投资决策的动态规划法。然后给出了该投资的最优临界值、最大机会价值以及不同波动率下的这两个值的变化趋势。研究表明:该“孪生”股票价格的条件异方差使得最优临界值和最大机会价值对波动率的敏感程度不同——当波动率增大时,上述两个值虽然都增加,但增加的程度不同;当波动率增大到一定程度时,这两个值增加的程度都明显提高。因此,将波动率纳入光伏发电投资决策分析中有助于提高决策质量,减少企业损失。     

中国股市短期波动的非线性ARCH效应显著性分析及参数估计

中国股票市场受政治或其它因素的影响有自己的波动特点。本文利用拉格朗日乘子,选择不同的频率检验中国股票市场短期的ARCH效应的显著性,并用FGLS四步法进行非线性估计。结果表明:中国股票市场在选择适当的频率下,短期的ARCH效应是显著的,对于研究中国股票市场的成熟度,预测股票短期投资风险,规范中国股票市场发展等具有重要的意义。     

Upper bounds for discrete moments of the derivatives of the riemann zeta-function on the critical line*

Assuming the Riemann hypothesis, we establish upper bounds for discrete moments of the Riemann zeta-function and its derivatives on the critical line. Moreover, we express continuous moments of the Riemann zeta-function and its derivatives in terms of these discrete moments. This allows us to give conditional upper bounds for $ {\int_0^T {\left| {{\zeta^{(l)}}\left( {{{1} \left/ {2} \right.} + {\text{i}}t} \right)} \right|}^{2k}}{\text{d}}t $ , where l and k are nonnegative integers.     

双变量参数联合函数的条件均值和条件方差(英文)

联合函数是指连接单变量边际分布的多变量函数.联合函数由Sklar(1959)在概率测度空间的内容时引入的.本文主要对边际分布是标准正态分布函数U(0,1)的Farlie-Gum-bel-Morgenstern和Gumbel-Hougaard这两个双变量参数联合函数进行研究,我们得到了他们密度函数的基本性质并导出了他们的条件均值和条件方差.另外,本文还给出了不同参数的条件均值和条件方差的相应图示,并进行了对比和解释.     

Some properties of record values coming from the geometric distribution

The independence between spacings of record values and of individual record values is well known when the records come from a geometric distribution. Here we examine the form a function of two record values must have if we require independence from a lower record value. Also similar questions are examined in relation to conditional expectation and conditional variance.     

Quasi-stationary distributions for absorbing continuous-time denumerable Markov chains

The stationary conditional, doubly limiting conditional and limiting conditional mean ratio quasi-stationary distributions are given for continuous-time Markov chains with denumerable state space both in terms of the transition matrixP(t) and the infinitesimal, generatorQ.     

Multiple Imputation by Ordered Monotone Blocks With Application to the Anthrax Vaccine Research Program

Multiple imputation (MI) has become a standard statistical technique for dealing with missing values. The CDC Anthrax Vaccine Research Program (AVRP) dataset created new challenges for MI due to the large number of variables of different types and the limited sample size. A common method for imputing missing data in such complex studies is to specify, for each of J variables with missing values, a univariate conditional distribution given all other variables, and then to draw imputations by iterating over the J conditional distributions. Such fully conditional imputation strategies have the theoretical drawback that the conditional distributions may be incompatible. When the missingness pattern is monotone, a theoretically valid approach is to specify, for each variable with missing values, a conditional distribution given the variables with fewer or the same number of missing values and sequentially draw from these distributions. In this article, we propose the “multiple imputation by ordered monotone blocks” approach, which combines these two basic approaches by decomposing any missingness pattern into a collection of smaller “constructed” monotone missingness patterns, and iterating. We apply this strategy to impute the missing data in the AVRP interim data. Supplemental materials, including all source code and a synthetic example dataset, are available online.